The overall aim of this proposal is to create genetically altered repressors that can regulate transcription in animal systems. This grant describes how we plan to create repressors that (a) can recognize different operators and (b) can be induced with non-toxic inducers. The research proposed will also help to elucidate principles governing protein-DNA interactions. Building on our structural studies of the lac repressor, we have created directed libraries of repressor molecules by randomizing the residues making specific interactions with bases in the operator sequence. We will determine how alternate sequences can interact with specific bases using an in vivo assay. The diversity in the sequences of these modified repressors will be used to unravel the molecular basis for protein-DNA recognition. We also propose to create heterodimeric repressor that can bind more diverse operator sequences. Most repressors recognize operator sequences that are nearly symmetric. A heterodimeric repressor has the ability to recognize an operator with two distinct half sites and therefore bind specifically to virtually any DNA. Our final aim is to create repressors that can be induced more efficiently by both existing and novel inducers. We have created a library of mutants where the residues lining the ligand binding pocket have been randomized. We will search this library for mutant repressors that are inducible. These mutant repressors can then be combined with mutant DNA binding domains to create new genetic switches distinct from the wild-type lac repressor. In the future a modified lac repressor could be used to regulate a variety of promoters. Endogenous loci could be switched on and off to create reversible models of human disease and normal development.